agent-me
View on GitHubDistill your knowledge, memories, and decisions into an open-source, inspectable AI Agent Twin.
An open-source personal AI twin built with FastAPI and React. It retrieves from reviewable Markdown knowledge and uses sequential planner, researcher, critic, and writer roles to produce evidence-grounded answers with inspectable traces.
Use Cases
Build a personal AI twin grounded in reviewed knowledgeAnswer questions using cited personal evidenceInspect and evaluate multi-stage agent workflowsExplore an open-source architecture for personal AI memory
Built With
- Language
- TypeScript
- Frameworks
- FastAPI · React
Tags
AI agent twin · personal AI · multi-agent · memory · RAG · evidence grounding · inspectable traces · evaluation